Law360 is providing free access to its coronavirus coverage to make sure all members of the legal community have accurate information in this time of uncertainty and change. Use the form below to sign up for any of our weekly newsletters. Signing up for any of our section newsletters will opt you in to the weekly Coronavirus briefing.
Sign up for our Compliance newsletter
You must correct or enter the following before you can sign up:
Thank You!
Law360 (June 22, 2020, 5:28 PM EDT )
 |
| Juanita Beecher |
 |
| Joanna Colosimo |
 |
| Jon Geier |
Tens of millions of people have become unemployed; many millions more are working from home. The jobs of many of the recently unemployed are likely lost forever; employees who do return to their jobs will likely do so in waves, may return to an altered schedule, and generally will find themselves in a very different workplace.
Employers will face novel selection decisions when returning employees to work. How does an employer determine which employees will be required to return first? Can those employees elect to work from home? Are employers subject to claims of discrimination based on who returns to work first and who remains on unemployment?
Moreover, as economic fissures continue to impact large and small employers it is likely that many employees currently working from home eventually will be subject to a reduction in force. RIF selection decisions will present equally unique challenges. The evaluation criteria and selection of employees who have been working remotely and are being managed remotely will be very different than in any prior reduction scenario.
This article provides in-house employment lawyers and human resources professionals, most of whom are likely working from home, a catalog of the unique legal issues, risks and practical concerns they should consider as we all adjust to our post-pandemic shutdown world.
Determine the Work That Will Continue
The first question that must be answered is what work will continue to be done. Then it must be decided who will perform the continuing work.
The decision of who will be returned to work from unemployment, as well as who among the currently working from home employees will be retained, can be made in a number of ways, including but not limited to, eliminating an entire department, reducing the workforce by a fixed percentage, or via an assessment of the knowledge, skills and abilities of individual employees. Regardless of how management decides to fill or eliminate jobs, it is essential that employment counsel and HR be involved at every step of the process to provide guidance.
Decide Which Employees Will Be Retained
In order to select which employees will be returned to work or retained in an RIF, management must develop neutral, objective and business-related criteria. In-house counsel and HR should review the proposed selection criteria to ensure management has not included criteria that might disfavor a protected group.
Criteria that might have an adverse impact against protected groups such as potential and performance ratings should be analyzed carefully. On the other hand, given the uniqueness of our current situation, criticality of knowledge, skills and abilities may be among the most important criteria to be evaluated as employers restart or reduce their operations.
Once the selection criteria is developed and reviewed to ensure objectivity and job relatedness, the next step is for management to use the criteria to determine which employees will return to work and which employees will be laid off, in many cases from working at home.
Analyze the Proposed Selections
After the proposed list of employees is developed, employment counsel should direct HR to run attorney-client privileged adverse impact analysis to determine whether there would be adverse impact based on gender,[1] race/ethnicity,[2] age[3] or perhaps even disability status.[4] Veteran status[5] should also be considered if the employer is a federal contractor. There are many factors to consider when constructing the analysis.
Is the decision a positive or negative employment consequence?
Any type of selection decision that can have a negative, or positive, employment consequence to employees can be considered for an equal employment opportunity analysis. Employment decisions such as layoffs, RIFs, furloughs, reductions in work hours or forced demotions are all examples of a negative employment consequence.
In constructing an EEO analysis for a negative decision, the most favored group is the group with the lowest selection rate. Employment decisions such as promotion, realignment or reorganization, that give the employee more pay or organizational status are examples of positive employment outcomes. Which employees return to work first is also a positive outcome and so the EEO analysis would examine the most favored group as the group with the highest selection rate.
What is the unit of analysis?
When structuring an EEO selection analysis, negative or positive, it is paramount to understand the unit of analysis to be used. The unit of analysis is the entity that frames what is being analyzed.
For example, a unit of analysis could be at a department level, job group level, job title level or organizational functional level. Some questions to consider include the following: How are/were the RIF or other decisions determined at the organization? Was there a readily available organizational structure or function that was targeted for RIF? Or, was a certain percentage of employees targeted across organizational units or functions?
The decisional unit may be reflective of an organizational function, such as marketing and finance. The best way to approach an EEO analysis would be to mirror the selection decisions. Defaulting the unit of analysis to something that is easy in the HR information system is likely not the best way to proceed if the goal of the analysis is to determine if there are EEO implications in the employment decision-making process.
Another consideration would be to analyze the workforce overall, in addition to mirroring the decisional units in the analysis, to determine if there is any overall impact.
What protected classifications need to be evaluated?
In addition to understanding how to mirror the decisional units in the RIF analysis, it is important to consider which EEO characteristics should be evaluated. Race/ethnicity, sex and age are the most common protected groups evaluated due to various legal implications and also due to availability of this data for employees.
When evaluating race or ethnicity it is advised to examine all protected subgroups for potential impact (e.g., Black, Asian, Hispanic, etc.). For federal contractors, approaching the analysis under the framework of the Uniform Guidelines on Employee Selection Procedures,[6] in which the most favored group is compared to other groups, is a common practice. In some cases, the most favored group may be a traditional racial minority group.[7]
Finally, an organization must consider how they plan to evaluate age in their EEO RIF analysis. A typical outcome is to compare individuals who are 40 years of age and over to employees under 40 years of age to align with the Age Discrimination in Employment Act. One consideration is what to use for the age/birthdate cutoff: the date of the decision or the date of notification to the employee?
What statistical tools should be utilized?
After the decisional units, protected and most favored groups are established, there are many statistical methodologies that can be considered when constructing an EEO analysis. Statistical tools that can be used to measure group differences include a chi-square analysis; a Z-test, commonly referred to as the two standard deviation test; and a rank sum test.[8] A Mantel-Haenszel test may be used to detect an overall pattern across decisional units.[9]
How do you evaluate both practical and statistical significance?
There are pros, cons and preferences to each of the above statistical approaches, but what should be considered is pairing the statistical test with a measure of practical significance as well. There are several inherent problems with using statistical significance as the only justification for an indicator or a claim of discrimination.
First, if the sample size is large enough, a relatively small difference could equate to a statistical finding. Second, the interpretation of a statistically significant finding depends on the parameters of the study. As the limitations of a study increase, the meaningfulness decreases.
Third, the two standard deviation threshold — indeed any threshold — is arbitrary, based on whether one considers a 5% likelihood (p less than 0.05) as sufficiently rare to view the results as unlikely to happen by chance. Indeed, Ronald Fisher, one of the founders of significance testing, first proposed the use of the p less than 0.05 level out of convenience, with social scientists later adopting it as the standard threshold.[10]
Thus, coupling statistical with practical measures of significance is important. Measures of practical significance include comparing the difference in selection rates, examining shortfall or expected counts, or the four-fifths rule — also known as the impact ratio or 80% rule.
After the privileged analyses have been performed, employment counsel should provide feedback to management and may recommend changes to reduce risk of discrimination claims. As a result of the adverse impact analyses, employment counsel may also suggest changes in the criteria if it appears certain criteria is having an adverse impact on any protected group.
Is WARN triggered?
If the employer has 100 or more employees or is in a state with a state Worker Adjustment and Retraining Notification Act, employment counsel and HR need to determine if the timing of the layoff will trigger a requirement to provide the WARN Act's 60 days' notice to the affected employees.[11] Federal WARN is triggered if there is a plant closing or if there is a mass layoff from a single site of employment as defined by the act.
If the layoffs would trigger an employer to provide 60 days' notice, either the employer must provide the notice, determine whether the proposed layoff allows the employer to provide less notice under one of the act's affirmative defenses, or craft the layoff to avoid triggering its 60-day notice requirement.
The WARN-trigger calculation may be especially complicated in the current crisis when previously laid off employees are returning in waves, while at the same time work-from-home employees are being laid off. Again, working with employment counsel will be critical in making this calculation.
What are the requirements if releases are used?
Most employers now offer employees who are being laid off enhanced severance benefits if they agree to sign a release. In order to ensure these releases are enforceable, employment counsel should review the requirements of the Older Workers Benefit Protection Act.[12] The OWBPA requires the waiver of rights by the employee be knowing and willing.
To that end the release must:
- Be in writing and be understandable; specifically refer to Age Discrimination in Employment Act rights or claims;
- Not waive rights or claims that may arise in the future;
- Be in exchange for valuable consideration in addition to anything of value to which the individual already is entitled;
- Advise the individual in writing to consult an attorney before signing the waiver; and
- Provide the individual with at least 21 days and at least 45 days for group waiver to consider the agreement before signing.
In addition, if the employer plans an exit termination, there are additional requirements to be considered as well.
Other Practical and Implementation Considerations
After the analysis and legal risk considerations have been conducted, there are many other practical considerations an organization should consider:
Document the process used.
It is essential that employment counsel and HR have management document the process and criteria used to make the selection decisions.
Consider the effect on diversity efforts.
Many employers have spent a great deal of time, effort and resources — human and financial — increasing the diversity of their workforce. Depending on the criteria management uses to select employees, the RIF could have a negative impact on that diversity and can undo the hard work and progress that had been accomplished.
Although race and gender cannot be determining factors in selection decisions, employment counsel and HR can ensure the proposed criteria do not unduly impact recent hires where possible.
Communicate decisions to employees.
Practitioners should plan ahead, utilize scripts and coach HR employees on effective communication strategies for delivering the news to both impacted employees, but also the rest of the employee population. Ultimately, notifying employees with dignity and measured compassion will influence the morale and productivity for the rest of your employees.[13]
H. Juanita Beecher is counsel at Fortney & Scott LLC.
Joanna Colosimo is director of compliance and workforce analytics and a principal consultant at DCI Consulting Group Inc.
Jon Geier is a principal consultant at DCI Consulting.
The opinions expressed are those of the author(s) and do not necessarily reflect the views of the firm, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.
[1] Title VII of the Civil Rights Act of 1964 as amended, 42 U.S. C. §2000e et al.
[2] Id.
[3] Age Discrimination in Employment Act of 1967 as amended, 29 U.S.C. §621 et al.
[4] Title 1 of the Americans with Disabilities Act of 1990 as amended, 42 U.S. C. §121010 et al.
[5] Vietnam Era Veterans' Readjustment Assistant Act of 1974 as amended, 38 USC Sec. 4212 et al.
[6]Uniform Guidelines on Employee Selection Procedures, 42 U.S.C. 2000e-12 and 2000e-8(1978).
[7] See, OFCCP v. VF Jeanswear, 2011-OFC-00006, 2013 Westlaw 4406032, ALJ's Recommended Decision and Order Granting Defendant's Motion For Summary Judgment (Dep't of Labor Aug. 5, 2013).
[8] The Chi Square statistic is commonly used for testing relationships between categorical variables. A z-testis a statistical test where expected selection rates for groups are compared against actual selection rates for both groups. A rank sum analysis is a nonparametric statistic that can be an alternative to the two-sample t-test and is based on the order in which the observations between two groups fall.
[9] A Mantel-Haenszel can help assess whether impact occurred overall by examining individual analyses combined.
[10] Ziliak, F.T., & McCloskey, D.N. (2008). The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice, and Lives. University of Michigan Press.
[11] Worker Adjustment and Retraining Act, 29 U.S.C. §2100 et al.; 20 C. F. R. Part 639.
[12] Older Worker Benefit Protection Act, 29 U.S.C. §626 (f); 29 C.F.R. Part 1625.
[13] The Society for Human Resource Management (2020). Retrieved at https://www.shrm.org/resourcesandtools/tools-and-samples/how-to-guides/pages/conductlayofforrif.aspx.
For a reprint of this article, please contact reprints@law360.com.
